Wired Article on the GPL

Richard Watts rrw1000 at cl.cam.ac.uk
Fri Mar 31 10:18:14 UTC 2000


On Thursday 30 March 2000, John Cowan
<jcowan at reutershealth.com> wrote:

>"Matthew C. Weigel" wrote:
>
>> Ummm... yes, you can accept or reject the GPL, if I understand it correctly.
>> You either accept the terms of the license -- the restrictions placed on
>> distribution, for instance -- or you don't, and if you don't, you have no
>> legal recourse for distribution.
>
>You can "accept" the GPL by exercising your rights under the GPL in accordance
>with its terms, or you can "reject" the GPL by not exercising your rights.
>But this is not acceptance as understood in discussing contracts.  

 .. unless the GPL is a contract. I got mail last night from someone
whose name I'll publish if he wants me to (but since he didn't post to
the list I think I ought to preserve his privacy), making an analogy
which got me thinking (read: I changed my mind).

 Suppose A gives me a piece of software, X, and agrees to licence it
to me under the GPL. The GPL allows me to do a number of things, but,
critically, section 3 requires me to distribute source code with my
binaries - that's a consideration. It's clearly valuable.

 Now, if I choose to distribute my binaries to the public, I must also
choose to distribute them to the author of the software, since he is
(presumably) a member of the public (jurisdiction warning
.. :-()). In which case I must distribute my modified source code
back to him. This is consideration - it goes beyond a plain licence
to do something: not only _can_ I distribute the source, I _must_ -
and it's valuable, and the fact that it's contingent is entirely
irrelevant - people make contingent contracts all the time.

 Of course, the author also gets vicarious benefits from the perceived
greater reliability of the software he uses which is based on the
software he's written, even if none of it was actually distributed to
him (i.e. I write gcc, make it open-source, give it to X, who is
forced to distribute source. Y takes X's bugfixes, modifies the
compiler, and uses it to compile the software for my PDA, which then
costs less: X has given me a vicarious consideration. Kind of.). But I
don't think that will fly...

 As to contractability, a contract exists wherever both parties
intend to bind themselves. The author clearly intends to bind himself,
or he wouldn't've offered the GPL. The recipient can avoid binding
himself by rejecting the GPL, but if he accepts, must accept the
conditions of section (3), and we therefore have a contract.

 I have no idea whatever if this would fly, but it's an argument.

[snip]
>> If you choose to not accept the license, then you can't use the software.
>
>Definitely not true of GPLed software.

 I guess this depends on whether you've decided if there's
consideration or not. Either way, if you exercise the permission to
make object files, then don't distribute the source code with them
under section 3, you're arguably in breach of something, and if it
isn't contract, it must be copyright.

>  You may not copy, distribute, or
>create derivative works except under the terms of the GPL.  But you can use
>the software on one machine or many; you can study the software to determine
>how it does what it does, etc. etc.

 Surely  that depends on the legal scope of copying, which I don't think is
quite settled yet ?

[snip]



Richard.



More information about the License-discuss mailing list